[Time Tracking], [Rovo], [Apps for Jira], [Timetracker]
Between March 10 and 12, 2026, our STAGIL team set aside three days for one focused goal: build and validate a Timetracker Rovo Agent prototype.
This wasn’t about creating a finished product. It was about speed, learning, and testing whether the idea actually works in practice.
Why This Matters
Time tracking in Jira is necessary—but rarely smooth.
Finding the right issue, filling in details, handling tags or attributes, and making sure everything is accurate all take time and attention. Small friction points quickly add up.
With this hackathon, we wanted to explore a different approach:
Can a dedicated Rovo agent handle Jira and Timetracker use cases in a practical, usable way?
The Plan and Our Goal
From the beginning, we defined a clear and focused scope.
The agent should be able to:
- Find Jira work items based on user input
- Show recent issues where work was logged
- Create worklogs with all required details
- Support Timetracker-specific data like tags, attributes, and billing fields
- Ask for confirmation before logging time
- Clearly explain missing inputs or errors
- Display recent worklogs in a structured way
We also planned additional capabilities—such as reporting and activity-based recommendations—if time allowed.
What We Delivered in 3 Days
By the end of the hackathon, we had a working prototype that covers the full core flow.
It includes:
- Jira user and issue search (including JQL)
- Full worklog handling (create, update, delete)
- Support for Timetracker metadata (teams, tags, attributes)
- Reporting with filters, date ranges, and time zone handling
- Automated environment setup via a custom Forge CLI tool
In practice, this means:
-
finding the right issue
-
logging time with all required details
-
generating reports
all within one connected flow.
What Worked Well
End-to-end flow delivered fast
We quickly built a complete flow—from issue search to worklog creation and reporting. This helped us validate early that the approach works in real scenarios, not just in theory.
Reduced input friction
By normalizing inputs (e.g. handling comma-separated values and applying defaults), we made interactions more reliable and reduced common errors.
Faster setup and onboarding
The custom Forge CLI tool and templates significantly sped up deployment and made it easier to get started.
Time zone handling from the start
We included explicit time zone handling in API calls, which is essential for accurate reporting.
At the same time, user-specific time zones are not yet accessible and require further investigation.
What We Learned
The hackathon also showed where improvements are needed before moving forward:
-
Authentication needs to be seamless
API keys worked for the prototype, but should be replaced with built-in user-level authentication.
-
Rovo design takes iteration
Prompts and actions required multiple refinements due to current limitations.
-
AI can support development
Using AI for prompt engineering helped speed up iteration and improve results.
-
Responses need optimization
Current API outputs are not yet optimized for Rovo, which impacts efficiency and cost. More focused data structures are needed.
What’s Next
With the concept validated, the next step is to strengthen the foundation:
- More robust validation and error handling
- Standardized API communication
- Proper user-level authentication
- Improved prompt and action design
- Optimized responses for better efficiency
- A solution for user-specific time zone handling
Final Thoughts
In three days, we built a working prototype that covers the full core workflow—and even extends into reporting.
More importantly, we confirmed that this approach can support real use cases, including more complex elements like tags and attributes.
Now the focus shifts from validation to refinement.
The direction is clear.
Interested in Timetracker?
Timetracker is already a powerful solution for time tracking in Jira, with features that support real, day-to-day work.
At the same time, we’re actively working on new capabilities, with more improvements coming soon.